Documents
Resources
Learning Center
Upload
Plans & pricing Sign in
Sign Out

Weaknesses of the Indicator Variable Approach in Short-term Event Studies

VIEWS: 7 PAGES: 18

Savickas [2003, Journal of Financial Research 26 (2), 165-178] proposes a new method - an indicator variable approach for testing abnormal returns in short-term event studies and finds the superiority of this approach over seminal methods (i.e., the parametric standardized cross-sectional approach of Boehmer, Musumeci, and Poulsen, 1991; the nonparametric approach of Corrado, 1989) in terms of test power. Through large-scale simulations of several variations of the GARCH models, we find that when an indicator variable is included in the conditional mean portion of the model, the increased power observed is due to the fact that an overall event hypothesis is being tested. This overall event test has little ability to detect event effects on different days in the event window. Neither does the indicator variable in the conditional variance portion of the model have much effect on the test power. The indicator variable approach needs to be carefully applied and interpreted for future event studies. [PUBLICATION ABSTRACT]

More Info
									INTERNATIONAL JOURNAL OF BUSINESS, 13(3), 2008                         ISSN: 1083−4346



  Weaknesses of the Indicator Variable Approach
          in Short-term Event Studies

                      Ronald Bremera, Zhaohui Zhangb
                  a
                   Rawls College of Business, Texas Tech University
                                Lubbock, TX 79409
                              ronald.bremer@ttu.edu
        b
          College of Management, Long Island University – C.W. Post Campus
                              
								
To top